Empirical results

Overview

This report summarizes the empirical results from the paper “Using newspapers for textual indicators: which and how many?” by E. Andres-Escayola, C. Ghirelli, L. Molina, J.J. Pérez, and E. Vidal. In particular, here we present the constructed EPU indices and the impulse response analysis from the Bayesian VAR model. Please refer to the paper for specific details¹.

Prepared by E. Andres-Escayola

Table of contents

EPU indices

---

• Regional
• Individual

Baseline IRF results

---

• Press source comparison
• Press breadth comparison
• Additional results

EPU indices

---

Regional

Latin America

---

(Index)

Note: This figure shows the EPU indexes for Latin America based on the local (red line) and foreign press (black line) against the narrative of events associated with increases in policy uncertainty in Latin America. The Latin American region is defined as the following countries: Argentina, Brazil, Chile, Colombia, Mexico and Peru.

Individual

Argentina

---

(Index)

Note: This figure shows the EPU indexes for Argentina based on the local (red line) and foreign press (black line) against the narrative of events associated with increases in policy uncertainty in Argentina.

Brazil

---

(Index)

Note: This figure shows the EPU indexes for Brazil based on the local (red line) and foreign press (black line) against the narrative of events associated with increases in policy uncertainty in Brazil.

Chile

---

(Index)

Note: This figure shows the EPU indexes for Chile based on the local (red line) and foreign press (black line) against the narrative of events associated with increases in policy uncertainty in Chile.

Colombia

---

(Index)

Note: This figure shows the EPU indexes for Colombia based on the local (red line) and foreign press (black line) against the narrative of events associated with increases in policy uncertainty in Colombia.

Mexico

---

(Index)

Note: This figure shows the EPU indexes for Mexico based on the local (red line) and foreign press (black line) against the narrative of events associated with increases in policy uncertainty in Mexico.

Peru

---

(Index)

Note: This figure shows the EPU indexes for Peru based on the local (red line) and foreign press (black line) against the narrative of events associated with increases in policy uncertainty in Peru.

Spain

---

(Index)

Note: This figure shows the EPU indexes for Spain based on the local (Spanish—red line) and foreign (Anglophone—blue line) press against the narrative of events associated with increases in policy uncertainty in this Spain.

Baseline IRF results

---

Press source comparison

IRFs of GDP growth to EPU shocks

Latin America

---

(Percentage points)

Note: The figure depicts the median impulse response of GDP to a rise of one standard deviation in the EPU index in the Latin American (LATAM) region: the red, blue and black lines represent responses to EPU shocks based on local, foreign and all press, respectively. The EPU indexes for the LATAM region are constructed as the simple mean between the country-specific EPU indexes. Foreign sources are represented by both Anglophone and Spanish newspapers. Filled symbols indicate statistical significance within the 84%–16% credible set, while empty symbols represent not-significant estimates. The horizontal axis measures quarters since the shock.

Brazil

---

(Percentage points)

Note: The figure depicts the median impulse response of GDP to a rise of one standard deviation in the EPU index in Brazil: the red, blue and black lines represent responses to EPU shocks based on local, foreign and all press, respectively. Foreign sources are represented by both Anglophone and Spanish newspapers. Filled symbols indicate statistical significance within the 84%–16% credible set, while empty symbols represent not-significant estimates. The horizontal axis measures quarters since the shock.

Mexico

---

(Percentage points)

Note: The figure depicts the median impulse response of GDP to a rise of one standard deviation in the EPU index in Mexico: the red, blue and black lines represent responses to EPU shocks based on local, foreign and all press, respectively. Foreign sources are represented by both Anglophone and Spanish newspapers. Filled symbols indicate statistical significance within the 84%–16% credible set, while empty symbols represent not-significant estimates. The horizontal axis measures quarters since the shock.

Spain

---

(Percentage points)

Note: The figure depicts the median impulse response of GDP to a rise of one standard deviation in the EPU index in Spain: the red, blue and black lines represent responses to EPU shocks based on local, foreign and all press, respectively. Foreign sources are represented by Anglophone newspapers. Filled symbols indicate statistical significance within the 84%–16% credible set, while empty symbols represent not-significant estimates. The horizontal axis measures quarters since the shock.

Press breadth comparison

IRFs of GDP growth to EPU shocks

Brazil

---

(Percentage points)

Note: The figure depicts the median impulse responses of GDP to a rise of one standard deviation in the EPU index in Brazil. The dashed grey line depicts the median IRF of GDP to shocks in the EPU constructed based on all press (local and foreign) and the grey area depicts its corresponding 84%–16% credible set. Foreign sources are represented by both Anglophone and Spanish newspapers. All other lines represent the IRFs of GDP to shocks in the EPU index constructed considering only one local newspaper at a time. Filled symbols indicate statistical significance within the 84%–16% credible set, while empty symbols represent not-significant estimates. The horizontal axis measures quarters since the shock.

Mexico

---

(Percentage points)

Note: The figure depicts the median impulse responses of GDP to a rise of one standard deviation in the EPU index in Mexico. The dashed grey line depicts the median IRF of GDP to shocks in the EPU constructed based on all press (local and foreign) and the grey area depicts its corresponding 84%–16% credible set. Foreign sources are represented by both Anglophone and Spanish newspapers. All other lines represent the IRFs of GDP to shocks in the EPU index constructed considering only one local newspaper at a time. Filled symbols indicate statistical significance within the 84%–16% credible set, while empty symbols represent not-significant estimates. The horizontal axis measures quarters since the shock.

Spain

---

(Percentage points)

Note: The figure depicts the median impulse responses of GDP to a rise of one standard deviation in the EPU index in Spain. The dashed grey line depicts the median IRF of GDP to shocks in the EPU constructed based on all press (local and foreign) and the grey area depicts its corresponding 84%–16% credible set. Foreign sources are represented by Anglophone newspapers. All other lines represent the IRFs of GDP to shocks in the EPU index constructed considering only one local newspaper at a time. Filled symbols indicate statistical significance within the 84%–16% credible set, while empty symbols represent not-significant estimates. The horizontal axis measures quarters since the shock.

Additional results

Comparison of local vs foreign press responses for Latin America²

IRFs of GDP growth to EPU shocks

---

(Percentage points)

Note: The panels depicts the median impulse response of the specified variable to a rise of one standard deviation in the local press (red) and foreign press (turquese) EPU indexes, along with the corresponding 90%– 10%, 84%–16% and 68%–32% credible sets (marked by decreasing colour intensity). The horizontal axis measures quarters since the shock.

IRFs of portfolio capital flows to EPU shocks

---

(Percentage points)

Note: The panels depicts the median impulse response of the specified variable to a rise of one standard deviation in the local press (red) and foreign press (turquese) EPU indexes, along with the corresponding 90%– 10%, 84%–16% and 68%–32% credible sets (marked by decreasing colour intensity). The horizontal axis measures quarters since the shock. In this specification the financial variable is non-residential portfolio capital flows as percent of GDP.

IRFs of equity prices to EPU shocks

---

(Percentage points)

Note: The panels depicts the median impulse response of the specified variable to a rise of one standard deviation in the local press (red) and foreign press (turquese) EPU indexes, along with the corresponding 90%– 10%, 84%–16% and 68%–32% credible sets (marked by decreasing colour intensity). The horizontal axis measures quarters since the shock. In this specification the financial variable is quarter-on-quarter changes of equity prices.

IRFs of exchange rates to EPU shocks

---

(Percentage points)

Note: The panels depicts the median impulse response of the specified variable to a rise of one standard deviation in the local press (red) and foreign press (turquese) EPU indexes, along with the corresponding 90%– 10%, 84%–16% and 68%–32% credible sets (marked by decreasing colour intensity). The horizontal axis measures quarters since the shock. In this specification the financial variable is quarter-on-quarter changes of exchange rates vis-à-vis the US dollar.

---

1. For questions on this report contact me at erikignasi@gmail.com.↩︎

2. These additional results are not part of the published paper and therefore should be interpreted with caution.↩︎